54 The Differential Impact of Genetic Moderators on the Relationship Between Depression and Cognition

Objective:Depression has a well-established negative effect on cognitive functioning. Variations in the apolipoprotein e (APOE) and brain-derived neurotrophic factor (BDNF) genes likely contribute to this relationship. APOE4 and the BDNF Val66Met polymorphism are independently associated with late-life depression and cognitive dysfunction. The current study investigated the moderating effects of APOE4 and BDNFMet (i.e., the presence of the BDNF Val66Met polymorphism) on the relationship between depression and cognitive functioning in older adults.Participants and Methods:The sample included 103 older adults drawn from two clinical trials who were recruited from the VA Palo Alto Health Care System (VAPAHCS) and the Stanford/VA Alzheimer’s Disease Center. Depression was diagnosed using the Mini Neuropsychiatric Interview for the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV). The presence of an APOE4 and BDNFMet allele were dichotomized (i.e., yes/no) and determined using venipuncture. A comprehensive neuropsychological battery was used to assess attention (RAVLT Trial 1, WAIS-IV DSF), processing speed (TMTA, SDMT, Stroop Word, Stroop Color), working memory (WAIS-IV DSB, DSS), visuospatial functioning (JLO), language (VNT), memory (RAVLT Delayed Recall, WMS-IV Logical Memory II), and executive function (TMTB, Stroop Color-Word). Separate moderation analyses were conducted with depression as the predictor and APOE4 or BDNFMet status as the moderator using the SPSS PROCESS macro v4.0. Age was a covariate for models with processing speed, memory, language, and executive function as outcome variables.Results:Participants were largely male (93%) and White (75%). Ten percent met criteria for depression, 26% were APOE4 carriers, and 32% were BDNFMet carriers. The overall model examining depression, APOE4, and memory was significant (p < .01, R2 = .14). Depression was associated with lower memory performance (p < .05), however, APOE4 was not a significant moderator (p > .05). Similarly, the overall model examining depression, APOE4, and language was also significant (p < .05, R2 = .10). While the direct effects of depression and APOE4 on language were nonsignificant (p > .05), there was a significant two-way interaction between APOE4 and depression (p = .03). The overall model with depression, BDNFMet, and memory was significant (p < .001, R2 = .18). While neither depression nor BDNFMet had significant direct effects on memory (p > .05), a two-way